May Review for F1000

Use of Abundance of One Species as a Surrogate for Abundance of Others.
Cushman SA, McKelvey KS, Noon BR, McGarigal K (2010) Conservation Biology Published Online: 7 Jan 2010 (early view) DOI: 10.1111/j.1523-1739.2009.01396.x

We have gone a little bit away from the biosecurity and invasive species theme this month, as we felt this paper was a very important more general paper.  There is a long history of the use of surrogate 'indicator' species in conservation management and scientific research that has never seemed entirely justified.  Others have challenged these ideas before, however, this paper provides a very clear message that caution should be used - proving rather than assuming that the abundance of one species may represent the abundance of another. 

Our Review 
Mark Lonsdale: Faculty of 1000 Biology, 11 May 2010 
 http://f1000biology.com/article/id/3184963/evaluation
Copyright F1000  

This stimulating article challenges the long-standing use of 'indicator' species concepts in conservation ecology. Such surrogate approaches, where a single species is assumed to represent others, have been used in many scientific studies and to justify management strategies.

This is a refreshing paper because it challenges the 'indicator' species concept in conservation biology, showing that for such a concept to be valid in the study and management of an ecosystem it 'must be demonstrated rather than assumed'. The authors carried out an extensive survey of birds in the Oregon Coast Range. They used two methods to group the birds into potential 'surrogate' groups: firstly, an 'a priori' grouping into life-history categories and, secondly, grouping 'empirically' using cluster analysis. The results for both methods show that some correlations can be found within groups that were defined both a priori and empirically. However, even when such correlations are found they do not explain a very high percentage of variance (generally below 20%), with the empirical grouping providing stronger patterns, largely because this grouping was based on the same dataset as the surrogacy test. Therefore, the empirical groupings represent the maximum variance explained by grouping and are probably unique to the dataset and location. 

Overall, the results give little support for the use of indicator species to infer the abundance of other bird species populations at two spatial scales (i.e. controlling for spatial dependence), as 'the lack of strong surrogacy is compelling' (pp 10). The authors were also able to highlight from their results that certain types of surrogacy are unlikely to exist, refuting concepts of 'guild-indicator' (use of a species as an indicator for an ecological guild, see Block et al., 1987 [1]) and 'management-indicator' species (use of a species as an indicator because its welfare is presumed to be an indicator of the welfare of other species in a habitat, see Landres, 1992 [2] -- an example is the use of the concept by the USDA for the management of the Sierra Nevada). 

The authors explain the failure of the indicator species concept with reference to ecological niche theory, stating that coexisting species are likely to differ in at least one critical niche dimension that will preclude them from acting as surrogates for one another. 

This article is not the first to highlight the importance of making assumptions about surrogate relationships in ecosystems. For example, Hurlbert (1997) questioned the popular use of the concept of 'keystone' species, showing that it couldn't be assumed that there exist species within a system that are somehow more 'important' or influential than others and that such species are unlikely to be easily identified[3]. Others have also cautioned the use of surrogate species, such as Wiens et al. (1998) [4], though not as strongly as this paper. 

The paper's conclusions are clear in stating that justifications for conservation and research emphasis on particular species should focus on the importance of the species itself, as it is unlikely that there will be sound justification for any one species' abundance to represent that of others.

References:
[1] Block, W. M., Brennan, L. A. and Gutierrez, R. J. (1987) Evaluation of guild-indicator species for use in resource management Environmental Management 11: 265-269 doi: 10.1007/BF01867205

[2]Landres, P.B. (1992) Ecological indicators: panacea or liability? Pages 1295-1318 in McKenzie, D.H., Hyatt, D.E. and Mcdonald, V.J. (eds) Ecological Indicators, Volume 2 Elsevier Applied Science, New York

[3]Hurlbert (1997) Functional importance vs keystoneness: Reformulating some questions in theoretical biocenology Austral Ecology 22 (4), Pages 369 – 382 doi: 10.1111/j.1442-9993.1997.tb00687.x

[4] Wiens, J.A., G.D., Holthausen, R. S., Wisdom, M.J. (2008) Using Surrogate Species and Groups for Conservation Planning and Management BioScience 58(3):241-252 doi: 10.1641/B580310